Bangladesh is entering one of the most dangerous moments in its recent political history. What began in July as a student movement against the country’s government-job quota system has expanded into a much broader confrontation with Prime Minister Sheikh Hasina’s government. After weeks of demonstrations, deadly clashes and a severe security crackdown, protesters are now demanding Hasina’s resignation rather than merely reform of the quota system. At least dozens of people were reported killed today alone as protesters, police and government supporters clashed across the country, adding to a death toll already in the hundreds since unrest began. The government has imposed restrictions including internet shutdowns and curfews, while the scale of mobilisation suggests the original policy dispute has transformed into something fundamentally different. The most important question is therefore no longer whether Bangladesh’s quota system was well designed. It is how a government can convert a policy dispute into a crisis of political legitimacy by making the cost of conceding appear greater than the cost of repression until repression itself becomes the principal grievance.
The original dispute concerned reservations in Bangladesh’s highly desirable government employment system. A substantial share of civil-service positions had historically been reserved for particular groups, including descendants of people who fought in the country’s 1971 war of independence. Students argued that the system disproportionately benefited politically connected families and reduced opportunities for applicants competing through open examinations. In a country where secure public-sector employment is extremely valuable and youth employment opportunities are limited, the issue connected immediately with a much larger economic anxiety. Government jobs offer stable salaries, social status and security unavailable to many young graduates in the private labour market. A rule determining access to those jobs therefore was not a technical administrative question. It affected one of the principal pathways through which educated young Bangladeshis imagine upward mobility.
This is the first reason the protests expanded so quickly: scarcity increases the political value of allocation rules. If high-quality employment were abundant, losing access to one category of government jobs would impose a smaller expected cost. When opportunities are scarce, every rule determining who receives them becomes politically consequential. A quota does not merely redistribute positions between two groups; it changes the probability that years of education eventually produce secure employment. Students therefore interpreted the system through fairness as much as economics. They were not simply asking how many jobs existed. They were asking whether access to those jobs depended primarily on merit or inherited political categories.
The economic logic resembles competition over university admissions or affirmative-action systems elsewhere, but Bangladesh’s liberation-war quota carries an additional historical dimension. The 1971 war is foundational to Bangladeshi national identity, and Hasina’s Awami League derives substantial political legitimacy from its association with the independence movement and with her father, Sheikh Mujibur Rahman, the country’s founding leader. Reserving jobs for descendants of freedom fighters therefore connects contemporary economic distribution with historical political legitimacy. Critics can consequently perceive the quota not simply as affirmative action but as a mechanism through which the political networks associated with the ruling party reproduce advantages across generations.
This creates the same historical-capital problem South Africa confronted with the ANC. Political movements that play central roles in national liberation acquire legitimate historical prestige. The danger begins when historical contribution becomes a claim over contemporary political or economic allocation. A citizen born decades after independence can respect the people who fought for the country while asking why ancestry should determine access to a government position today. Historical gratitude and present-day distributive justice are different questions.
The Awami League’s political dominance makes that distinction especially sensitive. Hasina has governed continuously since 2009, overseeing a period in which Bangladesh achieved substantial economic progress, particularly through garment manufacturing, infrastructure and improvements in several human-development indicators. The country transformed from an international symbol of extreme poverty into one of South Asia’s most significant manufacturing economies. Those achievements matter and help explain why Hasina maintained genuine support for years. Yet economic development and political liberalisation have moved in different directions. Opposition parties have faced increasing pressure, critics have been arrested and the January 2024 election occurred after the Bangladesh Nationalist Party boycotted the vote, producing another overwhelming Awami League victory in a contest the United States and others criticised as insufficiently free or fair.
Bangladesh therefore illustrates a recurring political-development problem: performance legitimacy can coexist with declining procedural legitimacy. A government delivers growth, infrastructure and poverty reduction, giving citizens rational reasons to value continuity. Leaders then begin interpreting successful performance as evidence that political constraints are obstacles rather than safeguards. Opposition appears irresponsible, media criticism appears destabilising and protest appears to threaten a developmental project that the government considers itself uniquely capable of managing.
The stronger the performance record becomes, the easier this argument is to make internally. Why risk political instability when the current government is delivering growth? Why allow disruptive opposition organisations to threaten investment? Why tolerate demonstrations that block roads and factories? Development becomes a justification for political concentration.
The problem is that economic competence cannot substitute permanently for political feedback. Governments make mistakes. The more political authority becomes concentrated, the fewer institutional channels remain through which citizens can communicate dissatisfaction before it becomes mass protest. Competitive elections, independent media, opposition parties and civil society are not merely liberal decorations around development. They are information systems. They tell governments where policies are failing and allow anger to be processed before it accumulates.
Bangladesh’s quota crisis demonstrates what happens when those feedback mechanisms weaken. A relatively specific policy grievance became the focal point for much broader frustration because students possessed limited confidence that ordinary political institutions would resolve it. Protest therefore became the mechanism through which information travelled upwards.
The government’s response then changed the nature of the protest itself. Security forces confronted demonstrators, violence escalated and large numbers of people were killed. Once that happens, the political utility function changes. A student who originally cared principally about employment quotas now has another grievance: friends, classmates or fellow citizens have been killed or arrested. The government can modify the quota and still fail to resolve the movement because the protest’s objective has become endogenous to the government’s response.
This is the repression-mobilisation paradox in its clearest form. Coercion raises the cost of participating in protest, which should theoretically reduce mobilisation. It simultaneously increases anger, solidarity and the perceived moral importance of participation. Which effect dominates depends on how citizens interpret repression. Limited enforcement can deter participation. Highly visible violence can transform fear into outrage, particularly when victims are students or other groups with broad social legitimacy.
The state’s use of lethal force therefore can change the composition of a movement. Parents who did not care strongly about quotas become concerned about their children. Lawyers mobilise around arrests. Opposition parties see an opportunity. Workers interpret the protests through their own economic grievances. Artists, academics and professionals join demonstrations because the issue becomes state violence rather than public-sector recruitment.
A single-issue movement becomes a coalition of grievances.
That transformation is politically dangerous because concessions addressing the original issue no longer dissolve the coalition. Bangladesh’s Supreme Court subsequently reduced the quota substantially, which should have addressed much of the students’ initial demand. Yet protests continued because the political equilibrium had already moved. Demonstrators increasingly demanded accountability for deaths, release of detainees and ultimately Hasina’s resignation.
This is why timing matters in crisis management. A concession made before repression can appear responsive. The same concession made after hundreds of deaths can appear insufficient or cynical. Governments therefore face a declining-return curve on compromise: the longer they wait, the larger the concession required to restore trust.
Political leaders frequently delay because early concession creates its own risk. If protesters learn that demonstrations produce immediate policy reversals, future groups may mobilise more readily. Governments therefore worry about establishing a precedent of weakness. Hasina’s government had incentives to demonstrate that street mobilisation could not dictate policy.
This is rational in a narrow repeated-game model.
It can become catastrophic when preserving deterrence against protest destroys legitimacy more quickly than protest itself.
The same escalation logic we have discussed between states therefore operates inside states. Government acts forcefully to deter future demonstrations. Protesters interpret force as evidence that greater mobilisation is necessary. Government interprets larger crowds as evidence that stronger coercion is required. Both sides escalate partly because each wants the other to stop escalating.
The difference is that the state possesses vastly greater coercive capacity.
That creates correspondingly greater responsibility for restraint.
A government controls police, intelligence services, prisons and potentially the military. Protesters may commit violence and should be held individually responsible where they do. The state’s institutional advantage means its decisions determine the scale of possible coercion much more heavily.
This is why state violence and protester violence cannot be evaluated simply by counting incidents symmetrically. The legal standards and capabilities differ. A citizen throwing a brick can commit a crime. A police unit firing on a crowd exercises public authority and therefore carries additional obligations involving necessity, proportionality and accountability.
Legitimacy comes from those constraints.
Without them, police become another armed faction rather than institutions of law.
The internet shutdowns demonstrate another dimension of state response. Governments facing unrest increasingly restrict mobile networks or social media because digital communication allows protesters to coordinate rapidly, document violence and distribute information outside official channels. From the government’s perspective, communications networks can facilitate disorder and misinformation. From protesters’ perspective, shutting them down prevents organisation and hides state abuses.
The technology therefore possesses dual political use. The same network coordinating peaceful demonstrations can coordinate violence. The same camera documenting police misconduct can distribute fabricated footage. Governments therefore face legitimate security concerns while possessing enormous incentives to use those concerns as justification for informational control.
Blanket shutdowns are especially problematic because they impose costs across the entire economy. Businesses lose payments and communications. Families cannot contact relatives. Journalists struggle to verify events. Emergency services can become less effective. The state therefore taxes millions of ordinary transactions to reduce the organisational capacity of a relatively small number of political actors.
This is another network externality of repression.
Digital infrastructure has become so integrated into economic life that shutting communications resembles closing roads to stop one demonstration.
The economic cost can be enormous.
For Bangladesh, this is particularly significant because the country’s development model depends heavily on integration into global supply chains. The garment industry accounts for the overwhelming majority of merchandise exports and employs millions of workers, many of them women. International buyers depend on predictable production, logistics and communications. Political instability therefore threatens one of the sectors responsible for Bangladesh’s economic transformation.
This creates a powerful contradiction for a government whose legitimacy depends partly on development. Repression intended to preserve political stability can undermine the economic stability supporting the government’s legitimacy. Factories close during unrest. Orders face delays. Foreign buyers diversify suppliers. Investors increase risk premiums.
A government can therefore win control of streets while losing confidence in boardrooms.
The relationship becomes particularly dangerous in export manufacturing because multinational companies possess alternatives. A garment brand sourcing from Bangladesh can shift some production towards Vietnam, India, Cambodia or other countries if disruptions become persistent. Capital and contracts are more mobile than workers.
Political instability therefore imposes asymmetric costs.
Workers remain in Bangladesh.
Orders can leave.
This is another reason political institutions should be understood as economic infrastructure. Predictable elections, peaceful protests and accountable policing reduce the risk premium attached to investment. Democracy may create visible political noise while producing a more predictable long-term environment because conflict is channelled through institutions rather than periodically exploding outside them.
Authoritarian stability can appear more efficient until it fails.
When failure occurs, suppressed grievances can emerge simultaneously.
This is the volatility trade-off of political systems. Democracies experience frequent small adjustments: governments lose elections, policies are reversed, journalists expose scandals, protesters demonstrate. Authoritarian or semi-authoritarian systems can suppress many of those fluctuations, producing apparent stability. The accumulated pressure may eventually produce a much larger discontinuity.
Bangladesh now appears to be experiencing precisely such a discontinuity.
Hasina’s long tenure created policy continuity and administrative coordination. It also reduced the number of credible mechanisms through which political succession could occur. The opposition BNP has been weakened and its leader Khaleda Zia has faced imprisonment and severe political constraints. Elections increasingly stopped functioning as plausible mechanisms of alternation.
That creates a succession bottleneck.
If citizens believe they cannot replace government through elections, every major political crisis begins carrying the question of regime change because there is no trusted scheduled mechanism for resolving dissatisfaction later.
In a consolidated democracy, protesters can demand a policy change while knowing an election will occur under credible rules. They do not necessarily need to remove the government immediately.
Where elections lack credibility, waiting becomes less attractive.
Street politics becomes existential.
This is why weakening electoral competition can paradoxically make governments less stable even while helping incumbents win elections. The government eliminates ordinary risks and creates extraordinary ones.
Political competition functions like a pressure-release mechanism.
Closing it does not eliminate pressure.
It changes where pressure accumulates.
The military now becomes crucial because Bangladesh has a long history of military involvement in politics, including coups and periods of military-backed rule. The armed forces therefore possess institutional weight far beyond ordinary security provision. As protests expand and police capacity becomes strained, political actors inevitably watch whether the military remains fully aligned with Hasina’s government, attempts mediation or refuses particular forms of repression.
This is where civil-military relations become decisive. A professional military should not determine who governs simply because streets become unstable. Elections and constitutional procedures should. Yet when civilian institutions lose legitimacy, militaries often justify intervention through the language of restoring order.
That is one of the most dangerous transitions available to a democracy.
Military intervention can stop immediate violence while creating a precedent that elected governments ultimately remain conditional on military approval.
Pakistan demonstrates how difficult that precedent becomes to reverse.
Bangladesh itself has experienced it before.
The strongest outcome therefore would be a political solution preserving civilian constitutional authority while restoring enough electoral credibility that future succession can occur normally.
That requires more than replacing one leader.
It requires rebuilding the mechanism through which leaders can be replaced.
This is why the protests’ demand for Hasina’s resignation, while understandable after the scale of violence, does not by itself provide a complete institutional programme. Who governs afterwards? Under what authority? For how long? Who organises elections? Which institutions guarantee that the next government does not reproduce the same concentration of power?
Revolutions and mass movements frequently become much clearer about whom they oppose than what institutional equilibrium should replace them.
Removing a leader is an event.
Building political competition is a process.
Bangladesh’s experience after previous periods of political upheaval demonstrates how quickly temporary arrangements can become contested. The country’s politics has been dominated for decades by rivalry between the Awami League and BNP, often accompanied by boycotts, strikes, arrests and allegations of electoral manipulation.
The problem therefore predates Hasina’s current crisis.
Bangladesh needs a political settlement where losing an election does not mean political elimination.
That is the same authoritarian-exit problem Venezuela faces this week.
If incumbents believe defeat means imprisonment and opposition leaders believe losing means permanent exclusion, every election becomes existential. Each side then has incentives to manipulate institutions while controlling them because it cannot trust the other side to exercise restraint later.
Democratic consolidation requires changing that repeated game.
Today’s winner needs to preserve rights it may need as tomorrow’s loser.
The Awami League would benefit from independent courts when it eventually enters opposition.
The BNP would benefit from neutral electoral institutions when it eventually governs.
Political actors need to build institutions from behind a veil of future uncertainty.
Bangladesh’s current system has struggled because parties often design rules while assuming they will remain on the favourable side of them.
Hasina’s government may now be discovering the danger of that assumption. Political dominance looked durable because the opposition was weak, elections were controlled and the state possessed substantial coercive capacity.
Then students mobilised around quotas.
This demonstrates the unpredictability of focal-point politics. Deep dissatisfaction can exist for years without producing coordinated opposition because citizens do not know whether others are willing to act. A seemingly narrow issue suddenly becomes a focal point around which previously disconnected grievances coordinate.
Nobody needs to plan the entire movement centrally.
Students protest quotas.
Government responds harshly.
Families become angry.
Opposition groups participate.
Workers join.
The grievance expands.
A new equilibrium forms faster than the state can adapt.
This is why authoritarian stability is difficult to measure. Low protest can mean high satisfaction.
It can also mean high coordination costs.
Observers cannot distinguish the two easily until something lowers those costs.
The quota protests did.
Hasina’s response lowered them further by giving different groups a shared grievance.
Repression can therefore solve the coordination problem for the opposition.
That is one of its greatest strategic risks.
The government’s economic achievements should still be acknowledged because the crisis does not erase them. Bangladesh’s GDP per capita increased dramatically during Hasina’s tenure, electricity access expanded and infrastructure projects transformed connectivity. The country became an important example of development driven partly by export manufacturing and women’s labour-force participation.
The political lesson is not that those achievements were fake.
It is that development cannot purchase permanent exemption from accountability.
Citizens update their reference points.
Once electricity access becomes normal, voters stop rewarding government simply for providing electricity and begin evaluating its reliability and price.
Once children attend school, parents evaluate educational quality.
Once graduates exist in larger numbers, they demand jobs matching their qualifications.
Development therefore creates new political demands through its own success.
This is the expectations revolution.
A government can become a victim partly of the aspirations its development helped create.
Bangladesh produced millions of educated young people.
Those young people now expect meritocratic employment, political voice and institutions matching a more modern economy.
The state cannot rely indefinitely on political structures designed for an earlier stage of development.
Economic modernisation therefore eventually generates pressure for institutional modernisation.
Not automatically.
China demonstrates that rapid development can coexist with durable authoritarian institutions.
But education, urbanisation and digital connectivity increase citizens’ organisational and informational capacity.
Governments need institutions capable of absorbing those capacities rather than treating them purely as threats.
The student movement is therefore partly a product of Bangladesh’s success as well as its failures.
A generation with enough education to compete for civil-service examinations is also a generation capable of scrutinising how those jobs are allocated.
Meritocracy becomes politically salient when people possess credentials whose value depends on fair competition.
This is why corruption and patronage become more destabilising as societies develop. Citizens invest more in education and therefore demand more predictable returns from that investment.
A political system allocating opportunity through loyalty rather than capability wastes human capital while generating resentment.
The quota debate therefore touches a fundamental development question: how does a state move from distributing opportunity through historical or political categories towards institutions capable of selecting talent efficiently while still correcting genuine structural disadvantage?
There is no simple answer.
Pure examination merit can reproduce inequalities created by unequal education.
Excessive quotas can weaken perceptions of fairness and efficiency.
Good systems balance representation, historical justice and capability while reviewing categories as social conditions change.
The key is that allocation rules need legitimacy among the population competing under them.
The Supreme Court’s reduction of the quota could have become the basis for such a compromise.
The violence surrounding the dispute has made that much harder.
Again, delayed institutional response increases the cost of settlement.
The immediate objective now should be preventing further loss of life. Protesters have announced a broad campaign of non-cooperation and are demanding Hasina’s resignation. Government supporters and security forces remain mobilised. Every additional death changes the political environment and makes compromise harder.
This is an escalation trap. The government may believe that restoring order is a prerequisite for political dialogue. Protesters may believe that dialogue is impossible until the government stops using force. Each side therefore makes de-escalation conditional on the other side moving first.
Breaking that trap requires an immediate, credible reduction in violence. Security forces should be instructed to protect life, avoid indiscriminate force and distinguish peaceful demonstrators from individuals committing attacks. Protesters and opposition leaders should likewise prevent attacks on police, public buildings, minorities and political opponents. Arrests should be reviewed transparently, detainees should have access to lawyers and independent investigations should examine deaths and allegations of abuse.
These measures are not concessions to one faction.
They are the minimum conditions for restoring a political arena in which concessions can matter.
The government should also recognise that a technical settlement on quotas is no longer sufficient. Any durable agreement must address the broader crisis of trust: accountability for killings, protection for peaceful assembly, restoration of communications, guarantees against collective punishment and a credible path towards political dialogue.
The opposition, for its part, must explain what follows resignation. A transitional arrangement cannot become an indefinite suspension of constitutional politics. Any interim authority would need a clear mandate, a limited timetable, civilian oversight and an agreed process for preparing credible elections.
The central challenge is to prevent the collapse of one form of political closure from producing another.
Replacing an elected government through mass pressure may be unavoidable if institutions have ceased to function, but it does not automatically create democratic institutions. A movement that defeats one concentration of power can reproduce it unless it establishes rules that constrain whoever wins next.
Bangladesh therefore faces two linked tasks.
The first is immediate: stop the killing and prevent the crisis from becoming a civil conflict or military intervention.
The second is structural: rebuild the institutions that allow citizens to change governments without risking national breakdown.
The first task requires restraint.
The second requires compromise.
Neither can be achieved through victory alone.
Hasina’s government once appeared politically unassailable because it combined economic performance, historical legitimacy, administrative control and a weakened opposition. Those sources of strength concealed a vulnerability: the system had become dependent on the assumption that citizens would remain politically fragmented and that coercion could contain any challenge that emerged.
The student movement disproved that assumption.
A quota dispute became a national crisis because it connected economic scarcity, historical privilege, generational frustration, state violence and the absence of credible electoral alternatives. The movement’s power did not come from the original issue alone. It came from the way the government’s response transformed that issue into a referendum on the political order.
That is the broader lesson.
Governments do not lose legitimacy only when they fail to deliver growth.
They lose legitimacy when citizens conclude that success cannot be questioned, policies cannot be corrected and leaders cannot be replaced.
At that point, even a narrow grievance can become a crisis of the state.
Bangladesh’s future will depend on whether its political actors understand that the present confrontation is not merely a contest over who controls the streets. It is a contest over whether political conflict will again be processed through institutions or continue to be settled through force.
If the country returns to repression without reform, the next crisis will likely be larger.
If it replaces one government without rebuilding competition, the same cycle will eventually return.
But if the current moment produces credible accountability, civilian constitutional authority, fair elections and a political settlement in which defeat is survivable, Bangladesh may yet convert a devastating crisis into an institutional turning point.
The choice is not between order and democracy.
It is between an order sustained by fear and an order sustained by legitimacy.
Only the second can last.
